These toxic elements can be bio-available to terrestrial and aquatic organisms, including crop plants, and could be further bio-accumulated via the food chain causing damage to humans.Since these metals cannot be degraded, current remediation approaches include excavation or capping, with a very high cost and damage to ecosystems.In many cases, these options are not economically feasible, when the contamination is very wide-spread as is the case of many contaminated farmlands and river beds.Compared to ex-situ remediation technologies, in-situ decontamination does not require excavation and transport of contaminated sediment and soil to off-site treatment or disposal facilities, thus it is generally a more practical and economical approach.Conventional insitu soil remediation technologies used for industrial sites contaminated with heavy metals include soil washing/flushing,nft hydroponic chemical immobilization, electro kinetic extraction, and phytoremediation.While these technologies may be appropriate for small scale remediation, they quickly become cost-prohibitive at larger scales.The cost of phytoremediation does not increase much with scale, but the accumulation of metals in the plants presents ecological risks and an eventual disposal cost.
Capping sediments essentially destroys habitat, and the capping may be removed during a large storm event, reexposing the contaminated media.Thus, there is an urgent need to find better methods to sequester heavy metals to reduce human and ecological risk and ensure better food security.Chelating agents, for instance, ethylene diaminetetraacetic acid , are widely used as extractive agents for heavy metals decontamination.Due to its strong metal chelating ability and low cost, EDTA has been used as a metal extraction agent in soil washing.However, soil washing can result in unintended mobilization of metals and other pollutants that can be more easily transported by groundwater, and EDTA itself can pose issues as secondary pollution.Thus, a suitable supporting material for EDTA and other chelating agents would minimize the potential unintended environmental implications.Previously we developed super-paramagnetic EDTA-functionalized nanoparticle adsorbents for water treatment, which were shown to remove a wide range of metal ions with high sorption capacity.To date, most nanoscale adsorbents have been applied to the decontamination of aquatic systems, while very few studies have investigated sediment and soil remediation.We have also demonstrated that nanoparticles can readily transport vertically into deeper soil, driven by gravity.Thus, we set out to develop a new type of high density nanoscale adsorbent, which can remove heavy metal ions during its downward transport, significantly reducing their bioavailability.For this study, we selected tungsten oxidenanoparticlesas the dense core, which is a relatively low-cost material with high density and low ecotoxicity, to develop the dense nanocomposites that can transport vertically through the porous medium.
We first report on the synthesis of EDTA-based Ligand DNPs.We then demonstrate the sorption capacity of Ligand DNPs for Cd2+ and Pb2+.Next, we evaluate the removal efficiency of Ligand DNPs for Cd2+ and Pb2+ in two different natural porous matrices.Finally, we report on the insitu remediation performance of Ligand DNPs for Cd2+ and Pb2+ during gravity-driven vertical transport in these media.The results demonstrate that Ligand DNPs can be applied for effective in-situ metal decontamination from soils and sediments.Pyridine and toluene were purchased from Alfa Aesar.triethoxysilanewas purchased from Sigma-Aldrich.Cadmium chloride anhydrous, lead chloride, ethylenediaminetetraacetic acid , and tris aminomethane were purchased from Fisher Scientific.Diethyl ether and sodium dihydrogen phosphate were purchased from Acros Organics.Standard Suwannee River natural organic matter was obtained from the International Humic Substances Society.A NOM stock solution was prepared by mixing a known amount of NOM with DIwater for 24 h.The pH of the stock solutions was adjusted to 8 with 0.1 M and 0.01 M NaOH and HCl.All chemicals were used as received, without further purification.All solutions were prepared with deionized water from a Barnstead NANOpure Diamond water purification system.Similar to our previous synthesis strategies, the core-shell Ligand DNPs were prepared in two steps.The WO3 nanoparticles were coated with APTES to form a silane polymer layer via hydrolysis reaction.Then, the surface was modified with EDTA by forming the amide bonds between the EDTA’s carboxylic acid groups and APTES coating’s amino groups.
WO3 nanoparticles were dispersed into 40 mL toluene in a flask.After mixing well, 0.4 mL APTES was added to attach an amino group to the WO3 particles.Then the flask was connected to a reflux system , which was then rotated at 30 rpmin a water bath at 90˚C, and refluxed for 2 h.After the solution cooled to room temperature , 2 mM EDTA and 60 mL pyridine were added.The mixture was again rotated at 30 rpm in a water bath at 90˚C in the reflux system for 2 h.After the solution cooled down to room temperature, 100 mL sodium bicarbonate was added to adjust pH to 8.0.Deionized water was used to rinse the particles twice and then decanted.The same rinsing procedure was performed twice with ethanol and then diethyl ether.The particles were dried at room temperature for 24 h, and stored in a capped bottle prior to use.Transmission electron microscopy images were obtained using a JEOL 1230 Transmission Electron Microscope operated at 80 kV.Scanning electron microscopy studies were performed on a FEI XL40 Sirion FEG Digital Scanning Microscope.The surface area and pore volume of Ligand DNPs were determined using a Micromeritics 3Flex Porosimeter.The functional groups of the Ligand DNPs were detected using a Fourier transform infrared spectrometer on a Nicolet iS 10 FT-IR Spectrometer.Two representative soils were used in this study, as examples of the application of Ligand DNPs to treat contaminated porous media.A grassland soil was collected from a flat, well drained grassy area at the Sedgwick Reserve in Santa Ynez, CA , and farmland soil was collected from a fallow field at an organic farm in Carpinteria, CA.The permit for collecting soil samples was authorized by Brenda Juarez.Soil properties can be found in the Supporting Information , in S1 Table in S1 File.Soils were air dried, sieved through a 2 mm mesh, and stored at 4˚C until use.The physicochemical properties of the sieved soil samples, including pH, texture, saturation percent, soluble salts, cation exchange capacity , conductivity, organic content, bulk density, and exchangeable NH4, NO3, K, and PO4, were characterized in our previous study, and available in the SI, shown as S1 Table in S1 File.Total W, Cd, and Pb concentrations of each soil were measured by digesting ~0.3 g soil samples in 10 mL 1:3 HNO3: HCl at 200˚C for 1.5 h in a microwave digestion system , followed by analysis via inductively coupled plasma mass spectroscopy.In order to simulate Cd or Pb contamination, 20 g of each type of soil were placed in 50 mL conical test tubes, mixed with 40 mL of 10 mg/L Cd2+ or Pb2+ solution on an end-over-end shaker with a speed of 70 rpm at room temperature for 7 days to ensure sufficient equilibration time.Then, the tubes were centrifuged at 10,000 rpm for 20 min to separate soil and the residual Cd2+ or Pb2+ solution, and the supernatant was collected for residual Cd2+ or Pb2+ concentration determination by ICP-MS.Soil saturated with Cd2+ or Pb2+ was preserved at 4˚C for the sorption studies.Air dried soil saturated with Cd2+ or Pb2+ was digested with 1:3 HNO3:HCl at 200˚C for 1.5 h in a microwave digestion system,nft system then analyzed via ICP-MS to determine the total Cd or Pb content.For batch sorption experiments, 20.0 mg of Ligand DNPs were first dispersed in 5 mL DI water, then mixed with 10 g of Cd2+ or Pb2+ contaminated soil , in 50 mL conical tubes at pH = 7.Then, these tubes were mixed on the end-over-end system with a speed of 70 rpm at room temperature for 7 days, to ensure sufficient equilibration time.Adsorption kinetics studies were carried out at the previously stated conditions but for a set amount of time, varying from 6-h, to 12-h, 24-h, 2-day, 3-day, and 7-day.The dosage of Ligand DNPs ranged from 3, to 5, 10, 15 and 20 mg to study the adsorption isotherms at pH 7.To evaluate the potential effect of NOM on the remediation performance of Ligand DNPs, the adsorption isotherms were conducted by first dispersing 3, 5, 10, 15 or 20 mg of Ligand DNPs in 5 mL NOM solution , then mixing with 10 g of each type of contaminated saturated soil for 7 days.
After mixing the Ligand DNPs with contaminated saturated soil for the specified time, the supernatant and soil were separated by centrifugation.Due to the high density, the immobilized heavy metals adsorbed by Ligand DNPs would be spun down.The treated soil was collected from the top layer to avoid the possible heavy metal binding Ligand DNPs, and then dried in an oven at 60˚C for 72 h, then digested for total Cd or Pb content analysis via ICP-MS.All experiments were conducted at ambient temperature.To investigate the decontamination capability of Ligand DNPs during gravity-driven transport through soil saturated with Cd2+ or Pb2+, first the contaminated soil was packed into 15 mL conical tubes.An opening with a diameter of 2 cm was made at the bottom of the tubes as the outlet of the system.Suspensions of 20, 40, 60, 80 and 100 mg of Ligand DNPs were dispersed in 5 mL DI water, respectively, and then evenly applied onto the top of each conical tube.After applying the Ligand DNP suspension and allowing the suspension to drip out, the soil columns were placed in horizontal position and air dried overnight then oven dried at 60˚C for 72 h.The dried soil was carefully removed from the conical tube in 3 cm segments, labeled top, middle, and bottom section.Sub-samples were weighed, then digested for Cd or Pb content analysis.The Ligand DNPs were mixed with the two Cd or Pb contaminated soils at pH 7 for 7 days to evaluate their isothermal sorption performance.As shown in Fig 2, the removal efficiency gradually increased as the dosage of Ligand DNPs increased, since this increases the number of active sites.The Ligand DNPs exhibited higher Cd or Pb removal efficiency when applied to farmland soil compared to grassland soil.As shown in S1 Table in S1 File, grassland and farmland soil exhibited significantly different physicochemical characteristics, particularly the organic and ionic concentrations.The CECof grassland soilis considerably higher than the CEC of farmland soil , which results in higher retention of cations, including Cd2+ and Pb2+, leading to much lower desorption from the contaminated soil to the soil-water interface.In addition, as shown in S1 Table in S1 File, the electrical conductivity was 142.1 μm/cm for farmland soil and 18.9 μm/cm for grassland soil, indicating a higher concentration of ionsin the leachate of farmland soil compared to grassland soil.Thus, there can be a higher soil-water interface concentration of Cd2+ or Pb2+ in farmland soil compared to grassland soil, which increases the accessibility and interaction between the active sites of Ligand DNPs and Cd2+ or Pb2+.In both Cd2+ and Pb2+ contaminated soil remediation scenarios, Ligand DNPs achieved higher removal efficiencies on contaminated farmland soil than grassland soil.Ligand DNPs exhibited higher removal efficiencies of Pb2+ from both farmland and grassland soils compared to Cd2+, which agrees with the sequence of their EDTA complex formation constants : 18.04 for Pb2+ and 16.46 for Cd2+.It suggests that the complexation between Pb2+ or Cd2+ and the EDTA-functionalized surface is the dominant removal mechanism.The time-dependent removal of Pb2+ or Cd2+ by Ligand DNPs in contaminated soil was evaluated in batch studies, as shown in Fig 3.Ligand DNPs showed quick removal of Pb2+ in contaminated farmland soils, with over 75% of maximum removal efficiency achieved in the first 6 hours, and a minor increase from 1 to 7 days, when Pb2+ in contaminated grassland soils were treated with Ligand DNPs.Thus, the sorption equilibrium of bio-available Pb2+ with Ligand DNPs can be rapidly reached within 1–2 days, with mixing, in both farmland and grassland soils.Similar removal performance was observed when applying Ligand DNPs for Cd2+ soil remediation, as over 70% of the maximum removal efficiency could be achieved in the first 6 hours for both soils.However, it took up to 3 days of mixing to achieve Cd2+ sorption equilibrium , suggesting Ligand DNPs exhibit a faster removal rate for Pb2+ than Cd2+, which is due to the stronger binding constant with EDTA.NOM concentration in the soil typically ranges from 0.5% to 5%.In the current study, the original grassland soil had a higher organic content than the farmland soil , showing a relatively wide range of organic content.In addition, soluble NOM can interfere with, or compete for, metal cation sorption.In order to evaluate the effect of soluble NOM on the removal of Pb2+ or Cd2+ using Ligand DNPs, an extra 1% NOM was spiked into the Pb2+ or Cd2+ contaminated soils.